Merge pull request #26 from MiroMindAI/test/yfd

refactor(config): simplify config and script naming

Author: BinWang28

config/benchmark/browsecomp-en-200.yaml

@@ -14,6 +14,8 @@ execution:
   max_tasks: null      # null = no limit, or specify a number
   max_concurrent: 5    # Number of parallel tasks
   pass_at_k: 1         # Number of attempts per task
+  max_retry: 5
+  exceed_max_turn_summary: true
 
 # OpenAI API key for evaluation (required for browsecomp since it has ground truth)
 openai_api_key: "${oc.env:OPENAI_API_KEY,???}"

…nt_rollback_new_tools_toolblacklist.yaml …agent_browsecomp-en-200_mirothinker.yamlconfig/fangda_agent_browsecomp-en-200_mirothinker_single_agent_rollback_new_tools_toolblacklist.yaml renamed to config/fangda_agent_browsecomp-en-200_mirothinker.yaml config/fangda_agent_browsecomp-en-200_mirothinker_single_agent_rollback_new_tools_toolblacklist.yaml renamed to config/fangda_agent_browsecomp-en-200_mirothinker.yaml

@@ -7,7 +7,7 @@
 # Configuration parameters
 NUM_RUNS=3
 BENCHMARK_NAME="browsecomp-en-200"
-AGENT_SET="fangda_agent_browsecomp-en-200_mirothinker_single_agent_rollback_new_tools_toolblacklist"
+AGENT_SET="fangda_agent_browsecomp-en-200_mirothinker"
 MAX_CONCURRENT=50
 
 # Set results directory with timestamp